Update README.md
Browse files
README.md
CHANGED
@@ -11,12 +11,12 @@ configs:
|
|
11 |
This is the benchmark for CI builds repair task as part of the
|
12 |
๐๏ธ [Long Code Arena benchmark](https://huggingface.co/spaces/JetBrains-Research/long-code-arena).
|
13 |
|
14 |
-
> ๐ ๏ธ CI build repair: given the logs of a failed GitHub Actions workflow and the corresponding repository shapshot,
|
15 |
> repository contents in order to make the workflow pass.
|
16 |
|
17 |
All the data is collected from repositories published under permissive licenses (MIT, Apache-2.0, BSD-3-Clause, and BSD-2-Clause). The datapoints can be removed upon request.
|
18 |
|
19 |
-
To score your model on this dataset, you can use [**CI build repair benchmark**](https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-
|
20 |
|
21 |
## How-to
|
22 |
|
@@ -32,7 +32,7 @@ To score your model on this dataset, you can use [**CI build repair benchmark**]
|
|
32 |
```
|
33 |
from datasets import load_dataset
|
34 |
|
35 |
-
dataset = load_dataset("JetBrains-Research/lca-ci-
|
36 |
```
|
37 |
|
38 |
Note that all the data we have is considered to be in the test split.
|
@@ -46,11 +46,9 @@ followed by the commit that passes the workflow successfully.
|
|
46 |
|
47 |
Note that, unlike other ๐๏ธ Long Code Arena datasets, this dataset does not contain repositories.
|
48 |
|
49 |
-
* Our [CI builds repair benchmark](https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-
|
50 |
-
their model to
|
51 |
-
|
52 |
-
runs
|
53 |
-
for all the datapoints.
|
54 |
|
55 |
### Datapoint Schema
|
56 |
|
@@ -60,7 +58,7 @@ Each example has the following fields:
|
|
60 |
| Field | Description |
|
61 |
|---------------------|------------------------------------------------------------------------------------------------------------------------------|
|
62 |
| `contributor` | Username of the contributor that committed changes |
|
63 |
-
| `difficulty` | Difficulty of the problem (assessor-based.
|
64 |
| `diff` | Contents of the diff between the failed and the successful commits |
|
65 |
| `head_branch` | Name of the original branch that the commit was pushed at |
|
66 |
| `id` | Unique ID of the datapoint |
|
@@ -83,7 +81,7 @@ Each example has the following fields:
|
|
83 |
```
|
84 |
{'contributor': 'Gallaecio',
|
85 |
'diff': 'diff --git a/scrapy/crawler.py b/scrapy/crawler.py/n<...>',
|
86 |
-
'difficulty': '
|
87 |
'head_branch': 'component-getters',
|
88 |
'id': 18,
|
89 |
'language': 'Python',
|
|
|
11 |
This is the benchmark for CI builds repair task as part of the
|
12 |
๐๏ธ [Long Code Arena benchmark](https://huggingface.co/spaces/JetBrains-Research/long-code-arena).
|
13 |
|
14 |
+
> ๐ ๏ธ CI build repair: given the logs of a failed GitHub Actions workflow and the corresponding repository shapshot, repair the
|
15 |
> repository contents in order to make the workflow pass.
|
16 |
|
17 |
All the data is collected from repositories published under permissive licenses (MIT, Apache-2.0, BSD-3-Clause, and BSD-2-Clause). The datapoints can be removed upon request.
|
18 |
|
19 |
+
To score your model on this dataset, you can use [**CI build repair benchmark**](https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-builds-repair/ci-builds-repair-benchmark).
|
20 |
|
21 |
## How-to
|
22 |
|
|
|
32 |
```
|
33 |
from datasets import load_dataset
|
34 |
|
35 |
+
dataset = load_dataset("JetBrains-Research/lca-ci-builds-repair", split="test")
|
36 |
```
|
37 |
|
38 |
Note that all the data we have is considered to be in the test split.
|
|
|
46 |
|
47 |
Note that, unlike other ๐๏ธ Long Code Arena datasets, this dataset does not contain repositories.
|
48 |
|
49 |
+
* Our [CI builds repair benchmark](https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-builds-repair/ci-builds-repair-benchmark) clones the necessary repos to the user's local machine.
|
50 |
+
The user should run their model to repair the failing CI workflows, and the benchmark will push commits to GitHub,
|
51 |
+
returning the results of the workflow runs for all the datapoints.
|
|
|
|
|
52 |
|
53 |
### Datapoint Schema
|
54 |
|
|
|
58 |
| Field | Description |
|
59 |
|---------------------|------------------------------------------------------------------------------------------------------------------------------|
|
60 |
| `contributor` | Username of the contributor that committed changes |
|
61 |
+
| `difficulty` | Difficulty of the problem (assessor-based. 1 means that the repair requires only the code formatting) |
|
62 |
| `diff` | Contents of the diff between the failed and the successful commits |
|
63 |
| `head_branch` | Name of the original branch that the commit was pushed at |
|
64 |
| `id` | Unique ID of the datapoint |
|
|
|
81 |
```
|
82 |
{'contributor': 'Gallaecio',
|
83 |
'diff': 'diff --git a/scrapy/crawler.py b/scrapy/crawler.py/n<...>',
|
84 |
+
'difficulty': '2',
|
85 |
'head_branch': 'component-getters',
|
86 |
'id': 18,
|
87 |
'language': 'Python',
|